Addressing bandwidth contention in SMT multicores through scheduling

Abstract

To mitigate the impact of bandwidth contention, which in some processes can yield to performance degradations up to 40%, we devise a scheduling algorithm that tackles main memory and L1 bandwidth contention. Experimental evaluation on a real system shows that the proposal achieves an average speedup by 5% with respect to Linux. 
DOI: 10.1145/2597652.2600109

1 Figure or Table

Topics

  • Presentations referencing similar topics